Training in progress, step 1000

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-tiny",
+  "_name_or_path": "openai/whisper-medium",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,
@@ -13,17 +13,17 @@
   ],
   "bos_token_id": 50257,
   "classifier_proj_size": 256,
-  "d_model": 384,
-  "decoder_attention_heads": 6,
-  "decoder_ffn_dim": 1536,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
   "decoder_layerdrop": 0.0,
-  "decoder_layers": 4,
+  "decoder_layers": 24,
   "decoder_start_token_id": 50258,
   "dropout": 0.0,
-  "encoder_attention_heads": 6,
-  "encoder_ffn_dim": 1536,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
   "encoder_layerdrop": 0.0,
-  "encoder_layers": 4,
+  "encoder_layers": 24,
   "eos_token_id": 50257,
   "forced_decoder_ids": [
     [
@@ -52,7 +52,7 @@
   "max_target_positions": 448,
   "median_filter_width": 7,
   "model_type": "whisper",
-  "num_hidden_layers": 4,
+  "num_hidden_layers": 24,
   "num_mel_bins": 80,
   "pad_token_id": 50257,
   "scale_embedding": false,
@@ -147,7 +147,7 @@
     50362
   ],
   "torch_dtype": "float32",
-  "transformers_version": "4.32.0.dev0",
+  "transformers_version": "4.33.0.dev0",
   "use_cache": false,
   "use_weighted_layer_sum": false,
   "vocab_size": 51865

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:350c027a0102cc2e923192a2d6670947c0e403a40dfdd71b6985c5b462e26ea8
-size 151099049
+oid sha256:9eecd6a1a9d66f82d4b80145e78a115c834a8f9039142a32c4b5fefe3e61c814
+size 3055754969

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7392e3e84b72730b0ed28b63667d30a7d2982d714830707dab651b19e18b073f
-size 4155
+oid sha256:ccbbaef0822ad5f6ceb0b75817a185b335a1bef711d46adb36251d376f2931fb
+size 4219